Post on 03-Feb-2018
transcript
70
Non-Agent Cognitive Alignment Frames in selected European languages Slávka Janigová, P.J. Šafárik University Košice
This article aims to summarize preliminary findings of our research in syntactic
typology focusing on the activation of a sample of cognitive chains in selected European
languages involving Non-Agent Arguments, subcategorized as Unintentional
Performer, Cognizer, Perceiver, and Emoter. The languages analysed in this paper
include English, Danish, Swedish, Italian, Spanish, Portuguese, Slovak, Lithuanian,
Bulgarian, Hungarian, Basque, and Georgian. The research was conducted via
syntactic datasheets filled out by bilingual academics, with English being a reference
language. The author intended to test an onomasiological approach to syntactic
typology, i.e. to take cognitive chains as the point of departure in typological analysis.
The research was aimed to survey the coding flags and their combinations that speakers
of the languages sampled employ to activate the selected Non-Agent cognitive chains;
to identify the preferred cognitive alignments frames in a particular language; and to
find if/how the cognitive alignment frames coincide with the genetic and morpho-
syntactic typological classes of the languages sampled. The results might also be
relevant for SAE-focused areal typology (SAE – Standard Average European), since
they may be used as an indicative benchmarking set to test the eurouniversal
hypothesizing that due to language contact Experiencer cognitive chains are
predominantly flagged as NOM/Subject in the SAE languages.
Keywords: onomasiological syntactic typology, cognitive chains, non-agent argument,
coding alignment, cognitive-alignment frames, Standard Average European
1. Some theoretical considerations
1.1 Onomasiological stance in syntactic typology – cognitive chains and coding alignment
The research whose outcome is presented in this paper was motivated by the author´s interest
in the findings of areal typology which is concerned with the study of languages whose speakers
are cohabiting in the same geo-cultural area (Körtvélyessy 2015). In context of the languages
of Europe, the term used to refer to this specific linguistic territory projected as distinct against
the rest of the world is Standard Average European (2015:2). Areal typology experimentally
ignores the genetic affiliation and morpho-syntactic language types and opens its research focus
to mutual contact influences of languages. The fact of sharing the same geo-cultural area by
speakers of various, even genetically or morpho-syntactically unrelated, languages invites an
expectation that it might play a role in the shaping and structuring of language users´ cognitive
experience, i.e. perception, conceptualization and structuring of the extra-linguistic reality
which is shared by them. Cohabiting language users might be hypothesized to influence each
other in perceiving the extra-linguistic world, and, consequently, some similarities might also
be foreseen in the ways these extra-linguistic categories are mentally analysed and shaped into
syntactically relevant abstractions and the ways they are projected in the surface syntagmas.
While most of the current areal typologists studying language parameters on the level of syntax
take a semasiological stance, based on the above expectation, in this research an
onomasiological approach was pursued, i.e. cognitive chains were used as the point of departure
for the analysis of the coding markers and their arrangement.
71
Thus, in an onomasiologically-grounded syntactic research the analysis starts from the
arrangement of cognitive roles, cognitive (valence) chains. Valence taken cognitively may be
understood as the chaining of the minimum obligatory accompaniments of verbal action to
render a cognitively complete set. These obligatory members of valence chains are referred to
as arguments. If valence is approached from a cognitive perspective, arguments are abstract
categories of syntactic meaning, generally referred to as Agent, Perceiver, Cognizer, etc., which
are projected on the surface of a clause as its clause elements Subject, Object, etc.
In order to activate particular argument chains language users choose from and combine
various formal flags, or markers, available in their concrete language. These may be both
explicit/surface and implicit. The former coding properties (Van Valin 2001: 34) include
inflection (of nouns, pronouns, articles, adjectives, verbs), Subject / Verb agreement,
pluripersonal concord, prepositions, aspect and tense verb contrasts, word order, prosody, etc.
A particular, generalized, arrangement of coding properties will be further referred to as coding
or surface alignment. The implicit means consist in the speaker´s ability to read a particular
arrangement of arguments as cognitively plausible based on a so called cognitive feasibility
check of a particular alignment of arguments that is inevitably performed when the overt
flagging fails (Janigová 2014: 19). If comparing the argument reading of sentences (1) and (2)
below, in English where the word order is used as explicit coding marker there is no ambiguity
as to the argument ordering of the surface segments, while in Slovak the case syncretism within
the neuter and masculine declension paradigms allows of two interpretation alternatives, and
the recipient must employ a cognitive feasibility test, or context, to identify the intended
alternative in a given situation:
(1) The car / is towing / a truck.
Agent / Action / Theme
versus
(2) Auto / ťahá / kamión.
Agent /Action/ Theme or
Theme /Action/ Agent
In Slovak, for example, out of 24 inflectional nominal paradigms only four paradigms formally
distinguish between NOM and ACC cases, which means that the case syncretism renders the
Slovak sentence more exposed to the implicit markers than the English sentence with its
grammaticalized word order (or the Slovak word order must be much more grammaticalized
than it is usually admitted).
1.2 Alignment polysemy and synonymy
While studying an interaction between the coding and argument alignments, there may be
observed cases of alignment polysemy and alignment synonymy. Instances of alignment
polysemy include those where one surface string (identical in terms of its lexical units, word
order and morpho-syntactic flagging) is capable of activating several cognitive chains. Thus an
English sentence:
(3) John /dropped / a pen. may activate two cognitive chains:
72
(3) a. Agent / Willed Action / Theme or
(3) b. Unintentional Performer/Unwilled Action/Specifier
Cases of alignment synonymy, on the other hand, occur when two formally distinct surface
strings activate the same cognitive chain. In Slovak, for example, there are two surface strings
to activate the chain: Emoter+Emotion+Emotion Focus:
(4) a. Ja mám rád tento obraz.
I.1.NOM.SG have.1.SG.PRS glad.M this picture.ACC.SG
‘I like this picture.’
b. Páči sa mi tento obraz.
like.3.SG.PRS REFL me.I.SG.DAT this picture.NOM.SG
‘I like this picture.’
Example (4)a. contains NOM-marked Emoter whereas in example (4)b. the Emoter receives
DAT case. As usual with lexical synonymy, here, too, synonymy is never full. There is always
a slight difference in the surface meaning which is due to a particular lexical nature of the verb
itself. Alignment synonymy and polysemy is obviously language-specific. So the Slovak
counterparts of the polysemantic alignment of the English sentence (3) would require two
distinct lexical verbs for (3)a. and (3)b. cognitive chains. This also means that apart from the
coding markers it is also the lexical items that are relevant in activating the intended argument
contrast (antonymic pairs of lexical verbs, valency-relevant aspect contrasts, adverbials, such
as intentionally inserted in (3)a.). The English examples (3)a. and (3)b. will correspond in
Slovak to (5)a. and (5)b., respectively, (two distinct lexical verbs and coding alignments are
employed):
(5) a. John-0 odhod-il per-o.
John-NOM.SG dropped-1.SG.PST pen-ACC.SG
Agent Willed Action Theme
(5) b. John-ovi spad-lo per-o.
John-DAT.SG dropped-3.SG.PST pen-NOM.SG
Unintentional Performer Unwilled Action Specifier
1.3 Cognitive alignment frames
In general, there are three major intransitive and monotransitive alignment types identified
cross-linguistically according to Haspelmath (2005): “accusative alignment, neutral alignment
and ergative alignment” (2005:1). These alignment types are based on the surface coding of the
verb’s elaborators. In the accusative alignment type the intransitive and transitive Subjects are
treated alike and kept distinct from the Objects. In the neutral alignment type all of them are
treated alike, i.e. they are identical in form, however, this ignores the fact that the Subjects and
Objects are actually not treated alike because there are other coding markers than just inflection
that distinguish them safely. Finally, in the ergative alignment it is the Subject of a transitive
verb that is kept distinct from the intransitive Subject and Object which are formally identical.
73
Onomasiologically, the elements of these surface/coding alignment types are perceived
as activators of argument chains. Hence, a mere identification of the coding alignment type
does not render the onomasiologically relevant information because of the one-to-many side of
the surface frames (argument synonymy and polysemy). There are languages that have distinct
case frames for Agent-focused argument chains and Non-Agent-focused argument chains (Van
Valin 2001: 28). Others may have special case frames even to activate finer distinctions within
the category of Non-Agent. The Avar, for example, is able to distinguish formally the Agent,
Perceiver/Cognizer and Emoter through the Ergative, Locative and Dative case frames,
respectively (Černý 1971: 47, 50). We therefore introduced the concept of cognitive alignment
frames or CA-frames which are a kind of interface between the argument chains and coding
alignment. They were also used to facilitate the onomasiological annotation of syntactic
structures. To exemplify the point, sentence (6) would be CA-frame annotated as follows:
(6) The car is towing a truck. > AGNOM-ThOBL
1.4 Diachronic perspective in genetic affiliations
Although in areal typology genetic ties among languages are neglected methodologically, for
the purposes of the present research they cannot be omitted. 9 of the languages in our sample
come from the common ancestor, i.e. the Proto-Indo-European. As for the Non-Agent argument
chains, they are associated with impersonal case frames in Proto-Indo-European. As Bauer
(2000) pointed out “[…] the impersonal verbs represent a pattern that was inherited from Proto-
Indo-European: this assumption is based on the consistency in structure – which typically
includes a third-person singular verb form – and also on the consistency in meaning. Impersonal
verbs in Indo-European languages typically convey three categories of meaning, (I)
meteorological conditions and natural events; (II) expression of emotions and feelings, and
physical experience, and (III) expression of modality.” (Bauer 2000:146). In category (II),
which is relevant for the present research, the structures typically showed the absence of Agent-
Nominative, and the presence of Oblique cases (DAT, ACC) to convey Experiencers of
emotional and physical conditions. It was in this category II where a shift started in Early Latin
from impersonal to personal structures (the former being diachronically original), whereas
personal instances were rather rare at that time, (although exact chronological order and direct
evidence is missing) (Bauer 2000:146). “[…] a switch has taken place in the history of Latin
during which the “Experiencer-Obliquus“ came to be replaced by a Subject-Nominative […]
the development of these verbs is in line with the increasing importance of the Subject-
Nominative in Indo-European languages.“ (Bauer 2000:128).
A similar shift as in Latin also occurred in other Indo-European groups. Its presence in
the Germanic group of Indo-European languages may be demonstrated by the shift occurring
in Old English and Middle English. OE had about 40 impersonal verbs distributed over
categories I through III. Category II was coded by the Oblique cases DAT and ACC. OE
impersonals survived to ME and for some time coexisted with their personal counterparts which
started to develop over time also under the influence of Scandinavian languages and, especially,
French. As Bauer (2000) pointed out, two important shifts occurred in respect of the genuine
OE verbs:
a) a shift from an impersonal verb governing a dative to an impersonal verb governing
a prepositional phrase: hit happened hem > it happened to him
74
b) a shift from impersonal verbs governing an accusative or a dative to a construction
consisting of a personal verb combined with a Subject-Nominative:
It reoweþ me > ich reowe hit – I regret it (Bauer 2000:128).
Thus the Oblique case frames employed to activate Experiencers are rather archaic forms that
survived from proto-language families stages of development till today even though they had
to compete with their Nominative rivals. The question is how universal this shift was in the
SAE area and if it has succeeded at all. Although diachronic considerations obviously exceed
the scope of the present research, and the author misses any data on the proto-case-frames for
Hungarian, Basque and Georgian, it seems to be worth pointing out the proto-prominence of
EXPOBL CA-frames in Indo-European languages in contrast to EXPNOM CA-frames.
2. Empirical part – a method and process of empirical analysis
2.1 Datasheets processing
Since, as was explained above, the empirical part of the research was approached from an
onomasiological perspective, 12 bilingual academics received datasheets containing a set of 12
sentences in English along with onomasiological charts. Respondents were asked to supply the
most natural counterparts of the English sentences in their second language (the list included
English, Danish, Swedish, Italian, Spanish, Portuguese, Slovak, Lithuanian, Bulgarian,
Hungarian, Basque, and Georgian). Since the sentences were decontextualized, rather simple
and word-order neutral, the functional sentence perspective and context issues were
disregarded. The onomasiological charts contained argument boxes into which respondents
were asked to insert their second language segments which they considered as activators of the
particular arguments. Segments were inserted in the boxes regardless of word order, while the
whole sentence was also supplied in a natural word order. If several variants of a particular
cognitive chain sounded equally natural, respondents were asked to supply both of them, i.e.
the occurrences of alignment synonymy as explained above. Respondents were also asked to
provide a brief morho-syntactic analysis using the Leipzig Glossing Rules. The morpho-
syntactic analyses of particular sentences varied due to individual preferences of respondents,
so the author sometimes had to tailor them to suit the research purposes of this study. The
analysis in Slovak is used to exemplify the point:
(1) SK (from the research corpus): I like that picture. >>>
i. Páči sa mi ten obraz.
ii. Mám rada ten obraz.
Emoter Emotion State Emotion Focus
I PRON.1.NOM.SG like V.1.SG that PRON picture N.OBL.SG
mi PRON.1.DAT.SG páči-0 V sa-REFL PART
like-3.SG
ten PRON obraz-0 N
picture-NOM.SG
75
(ja) PRON.1.NOM.SG
activated by-m- V Infl.1.SG
má-m AUX rad-a ADJ
have-1.SG glad-F
ten PRON obraz-0 N
picture-ACC.SG
The sentences received from respondents were subsequently distributed into tables, each
containing 12 language variants, and analysed as to the coding features, coding alignment and
CA-frames employed to indicate the Non-Agent Arguments – Experiencers (Emoter in (1) and
(2), Perceiver in (3), (4), Cognizer in (5), (6)), Unintentional Performer in (7), (9), (10), (11)
contrasted with Intentional Performer in (8) and Permitter in (12). The analysis is contained in
Appendix.
Here is the list of research sentences:
(1) I like that picture.
(2) I feel sorry.
(3) My leg hurts.
(4) I saw the man.
(5) It seems to me that John….
(6) I am interested in music.
(7) I dropped a pen. (unwilled)
(8) I dropped a pen (willed)
(9) I broke my leg. (unwilled)
(10) The car broke its axle.
(11) Centipedes grow their legs.
(12) John grows a beard in winter.
2.2 Typological ranking of languages in the sample
The languages in our sample all come from the Indo-European language family but for Basque
(no language family assignment), Hungarian (Ugro-Finnic language family) and Georgian
(Southern Caucasian). As for dominant word order, they are all SVO-type, with the exception
of Basque and Georgian which are SOV-type, and Hungarian that has no dominant word order.
Some of them are falling with the synthetic language type, some with the analytical. Table 1
shows the respective typologies.
Table 1: Typological characteristics of languages according to Millward (2012), Skalička
(2004) and Dryer (2013)
Millward´s Genetic
classification
Skalička´s morho-
syntactic typology
Dryer´s word
order typology
English Indo-European/Germanic Isolating SVO
Danish Indo-European/Germanic Isolating SVO
Swedish Indo-European/Germanic Isolating SVO
Italian Indo-European/Romance Isolating SVO
Spanish Indo-European/Romance Isolating SVO
Portuguese Indo-European/Romance Isolating SVO
76
Slovak Indo-European/Slavic Inflectional SVO
Lithuanian Indo-European/Slavic Inflectional SVO
Bulgarian Indo-European/Slavic Inflectional (verb
inflection only)
SVO
Hungarian Ugro-Finnic/Ugric Agglutinative No dominant
word order
Basque Basque Agglutinative SOV
Georgian Kartvelian
(Testelec 1998: 254)
Southern Caucasian
(Millward 2012: 47)
Inflectional/Agglutinative SOV
2.3 Case identification and marking
As for the formal marking of cases, according to WALS (Iggesen 2013), the languages sampled
have the following features (Slovak and Portuguese ranked by the author):
Table 2: Morphological marking of nominal cases
No morphological case
marking in nouns
Spanish, Italian, Bulgarian, Portuguese
2 cases distinguished by
the noun form
English, Danish, Swedish
6-7 cases distinguished
by the noun form
Georgian, Lithuanian, Slovak
10 or more cases
distinguished by the
noun form
Basque, Hungarian,
Since all the languages in our sample, except for Basque and Georgian, can be ranked with the
Nominative-Accusative languages, and also considering the onomasiological perspective of our
approach, we used the terms Nominative case and Oblique case to indicate the contrast between
the surface Subject and Object regardless of whether this contrast was activated by the
morphological form, position, preposition or agreement. Terminological labels, such as
Accusative and Dative, were used for languages which distinguish these cases by their
morphological case paradigms (here it should be pointed out that even though respondents for
languages, especially of the Romance language family, with no morphological case marking in
nouns insisted on that their languages ‘did not have cases’, they tended to identify some of the
Non-Agent entities as Datives or Accusatives, probably under the influence of the Latin
declension system and the case semantics associated therewith). In Ergative-Absolutive
languages (Basque and Georgian), the terms Ergative case, Absolutive case and Dative case
were used in respect of clause elements in surface paradigms where Subjects of intransitive
verbs and Objects of transitives were case-marked and cross-referenced in the agreement
elements of the verb identically, and differently from the Subjects of transitive verbs. The case
marking of transitive Subjects is called Ergative (in Basque mostly ending in -k), that of
intransitive Subjects and transitive Objects is called Absolutive (mostly zero inflection in
77
nouns) (Etxepare 2003:2-3). Actually the zero case, Absolutive “[…] is taken as the default
case, so that, unless otherwise stated, the absence of any case indication on a nominal means it
is in the absolutive case” (Hualde 2003:vii).
The pluripersonal verb agreement scheme (i.e. the verb´s morphological constituents
indicating concord with both the Subject and Object at the same time, by means of inflectional
prefixes, infixes, and suffixes) was a strikingly significant surface means that was employed to
activate a particular reading of argument chains in Basque and Georgian.
The author must confess that the inflectional decomposition of particular words was not
done thoroughly for all the inflected items in all the languages sampled. This was due to the
lack of the specific linguistic information necessary to show it. So where the meta-data were
limited, only the case was indicated as suggested by the respondent, which, however, fully
sufficed for the purposes of the present research.
2.4 Objectives of empirical analysis
The following is a list of research questions/objectives:
1. What coding features were employed by respondents to activate the selected Non-Agent
argument chains?
2. What was the distribution of coding features in the sample?
3. What Cognitive Alignment Frames were identified? What was the distribution of CA-
frames across the genetic and morpho-syntactic types of languages in the sample?
4. What were the major coding markers and CA-frames for intentional argument chains?
5. Testing the applicability of the Haspelmath’s EXPNOM SAE feature for the languages in
the sample.
3. Empirical findings
3.1 What coding features were employed by respondents to activate the selected Non-Agent
argument chains?
The following list includes about a dozen of recurring combinations of coding features that
were identified to activate the Non-Agent. In most cases joint employment of two or more
coding features was detected. Except for the coding alignments I (prominence of word order)
and VI (prominence of preposition) which are typical of highly analytical languages, majority
of the coding markers were inflectional, either nominal or verbal, or both. Moreover, the
inflection was hardly ever used as the sole indicator; it was accompanied with word order and
S-V concord, and in the case of Georgian and Basque with S-V-O pluripersonal concord
activated by the internal inflection of the verb. The list also includes reflexive particles,
Possessive forms and Genitive case as component of a noun phrase. The following is the list of
coding alignments (I-X) with certain subtypes, including examples from our research corpus:
I Nominative case indicated by word order / nominative form of pronoun 7 EN I dropped a pen. SVO I PRON.1.NOM.SG dropp-ed-0 V
drop-PST-1.SG
a ART
pen N.OBL.SG
IIa Nominative case indicated by nominal inflection/form + verbal inflection
+ S-V concord
78
1 PG Eu gosto deste quadro.
SVO eu PRON.1.NOM.SG
-o-V Infl.1.SG
gost-o V
like-1.SG
deste PREP+PRON
quadro N.OBL.SG
2
HUii
Kedvelem azt a képet.
‘I like that picture.’
(S)VO (én)
PRON.1.NOM.SG
activated by
-em-V Infl.1.SG
kedvel-em V
like-1.SG/INF
azt PRON
a ART
kép-et N
kép-ACC.SG
IIb Nominative case indicated by nominal inflection/form + verbal inflection + S-V
concord + OBL reflexive pronoun/particle 9 SK Zlomil som si
nohu.
‘I broke to
myself a leg.’
(S)VO (ja) PRON.1.NOM.SG
activated by
som-AUX
be-PRS.1.SG
si-REFL.DAT.SG
zlomi-l V
break-PST.M.SG
som AUX
be-PRES.1.SG
si-REFL.1.DAT.SG
nohu N.ACC.SG
III Absolutive case indicated by nominal form + verbal inflection + pluripersonal
S-V-O concord (in Basque and Georgian) 6
BASQ
Ni musican interesatuta
nago
SAV ni PRON.1.ABS.SG interesatuta V
nag-o AUX
nag-1.ABS.SG
musican N. LOC.SG
IVa Oblique case (DAT/ACC) indicated by nominal inflection/pronoun form +
verbal inflection + S-O concord 1 SP Me gusta esta foto. OVS me PRON.1.DAT.SG
-a-V Infl.3.SG
gust-a V
like-3.SG
esta PRON
foto N. NOM.SG
IVb Oblique case (DAT/ACC) indicated by nominal inflection/pronoun form +
verbal inflection + S-V-O pluripersonal concord (in Basque and Georgian) 1GEO me momc’ons is surati
‘That picture pleases
me.’
OVS me PRON.1.DAT.SG
activated by -m- DAT
INFL INFIX of V
under PLRPX
mo-m-c’on-s V
PRV-DAT.SG-like-
3.ABS.SG
is PRON
surati N. ABS.SG
IVc Oblique case (DAT/ACC) indicated by verbal inflection + S-V-O
pluripersonal concord (in Basque and Georgian) 3 GEO Pexi mt’k’iva.
(O)SV (me) PRON.1.DAT.SG
activated by m- prefix
of V under PLRPC
m-t’k’iv-a V
1.DAT.SG- t’k’iv-3.ABS.SG
pexi N.ABS.SG
V Oblique case (DAT/ACC) indicated by nominal + verbal inflection + reflexive particle 7 SP Se me cayo el
boligrafo
OVS me PRON.1.
OBL/.DAT.SG
se-REFL PART
cay-o V
fall-PAST.3.SG
el ART
boligrafo N.NOM.SG
VI Oblique case indicated by preposition + oblique nominal form 5 SW För mig det
verkar som att
John är glad.
OprepSVScl för PREP
mig PRON.1.
OBL.SG
verkar-0 V
seem-3.SG
det formal Subject
som att John är glad
(SUBORD FINITE
CLAUSE)
VIIa Possessive case form as determiner in word-order-based NOM nominal phrase 3DANi Mit ben gør ondt.
‘my leg does bad´
SVA mit PRON
POSS.1.SG
gør V.3.SG&PL
ondt ADJ
ben N.NOM. SG&PL
VIIb Possessive case form activated by nominal inflection in NOM nominal phrase
79
3HU Fáj a lábam.
VS -am-POSS.1.SG fáj V.3.SG a ART
láb-am N
NOM.POSS.1.SG
VIIc Possessive case form as determiner in word-order-based OBL nominal phrase 9EN I broke my leg.
SVO I PRON.1.NOM.SG
my-POSS.1.SG in OBL NP
broke-0 V.PST-1.SG leg N.OBL.SG
VIId Possessive case form as determiner in prepositional OBL nominal phrase 3
DANii
Det gør ondt i mit
ben.
‘it does bad in my
leg´
SVAA mit PRON
POSS.1.SG
gør V.3.SG
ondt ADJ
det formal Subject
i PREP
ben N.OBL. SG&PL
VIII Genitive case form as component of NOM nominal phrase (+ reflexive pronoun) IT Si è rotto l’asse
della macchina.
VS della PREP+ART
macchina N.GEN.SG
si REFL PART
é AUX
be-3.SG
rott-o
break-PST PTCP.M.SG
l’asse
ART+N.NOM.SG
IXa Ergative case indicated by nominal inflection + verbal inflection/auxiliary
+ S-V-O pluripersonal concord (in Basque and Georgian) 1
BASQ
i
Nik hori atsegin dut
‘I like that.’
SOV ni-k PRON
ni-1.ERG.SG
-t ERG INFL SUFFIX
of V under PLRPC
atsegin-V
d-u-t-ERG/ABS AUX
ABS.3.SG-u-ERG.SG
‘I have it’
hori PRON. 3.ABS.SG
IXb. Ergative case indicated by verbal inflection + S-V-O pluripersonal concord
4
GEO
Me davinaxe
k’aci.
SVO me PRON.1.ERG.SG
PLPRC
-v- ERG INFL INFIX of
V under PLRPC
da-v-i-nax-e V
PST-ERG.SG-VERS-nax-
ABS.SG
k’aci N.ABS.SG
X Hybrid
II/IV – nominative verb inflection + OBL nominal form 2 PGi Sinto-me
arrependido/a.
(S)VOA (yo) PRON.1.NOM.SG
activated by
-o-V INFL.1.SG
me PRON.1.ACC.SG
sent/sint-o V
feel-1.SG
arrependid-o/a
sorry-SG.M/F
context based
I+VIIa – word order + possessive form of pronoun in an OBL nominal phrase 9 EN I broke my leg.
SVO I PRON.1.NOM.SG
my-POSS.1.SG in OBL NP
broke-0 V.PST-1.SG leg N.OBL.SG
As can be seen, most of the coding alignments were further refined depending on the specific
coding markers involved. The Non-Agent arguments occupy either the Subject or Object
surface positions. Combinations involving the Possessive form of pronouns and the Genitive
activating Perceiver show quite a striking imbalance between the prominence of their cognitive
roles and their fully dependent surface position as either determiners within noun phrases
(VIIa,c,d) or their postmodifiers (VIII). The activator of Perceiver may even be reduced to a
morpheme level, as exemplified in Hungarian (VIIb). The Portuguese variant of sentence (2)
allowed of a combination of the Nominative and Oblique cases in one complex form (see the
hybrid coding alignment X above).
80
3.2 What was the distribution of coding features in the sample?
Table 3: Distribution of coding alignments used to activate the Non-Agent and two Agent
cognitive chains.
Sentence 1 2 3 4 5 6 7 8 Willed 9 10 11 12
Willed Language
English I I VIIa I VI I I I + ADVL VIIc VIIc I I
IVa
Danish I I VIIa I IVa I I I+
CAUS+
LEX
I VIIa I I+CA
US+
LEX VIIa IVa
Swedish I I VIIa I VI I I I+
CAUS+
LEX
I VIIa I I+CA
US+
LEX VIId IVa
Italian IVa Iva IVa IIa IVa IIa IVa IIa+CAUS V VIII+refl
VIII IIb+ CAUS
+REF
L
IVa IVa IVa IIb
Spanish IVa IIa IVa IIa IVa IVa V IIa+LEX IVa V+
extra
DAT
VI+
extra
ART DA
T
IIb+C
AUS+
LEX
Portuguese IIa X IVa IIa IVa IIa IVa IIa+LEX IIa IIa VI+
extra PRO
N DA
T
IIa+
CAUS IIa IIa+
CAUS+
LEX
Slovak IVa IVa IVa IIa IVa IIa IVa IIa+LEX IIb IIb IVa IIb+
LEX
IIa IVa IV IVa V IIb+C
AUS
Lithuanian IVa IVa IVa IIa IVa IIa IVa IIa+LEX IIb IVa IVa IIb
IVa
Bulgarian IVa IIa IVa IIa IVa IIa IIa IIa+LEX IIb+refl VIII
+refl
VI+
PRO
N POS
S in
NOM
NP
IIb+
LEX IIa
Hungarian VI IIa VIIb IIa IIa IVa IIa II+ ADVL VIIb VIII I+int
ensifier
IIa
IIa IIa
Basque IXa - IXa IXa IVc III IVb IXa+LEX
+AUX
VIIa IXa IXa IXa
IVc
Georgian IVc III IVc IXb IVc IVc IVc IXb+LEX IVc IVb IVb III
III IXb
81
3.3 What Cognitive Alignment Frames were identified? What was the distribution of CA-frames
across the genetic and morpho-syntactic types of languages in the sample?
From an onomasiological perspective, the coding alignments listed in 3.1 may be assigned to
three major and two minor Cognitive Alignment Frames, or CA-frames, (as was explained
above, the issue of case identification was approached onomasiologically in this research, i.e.
the case as an interface between the syntactic meaning and form regardless of whether activated
by the position, inflection, morphological form of a word, preposition, or concord). The cross-
projection of coding alignments and CA-frames is summarized in the Table 4 below and
distribution of CA-frames across the sample is summarised in Table 5.
Table 4: CA-frames, coding alignments and specification of coding markers
Cognitive Alignment
Frame (CA-frame)
Coding (surface)
Alignment
Coding Markers
A NANOM/ABS – SP/FOCOBL
Nominative/Absolutive Non-
Agent combined with
Oblique Specifier/Focus
I NOM+V+OBL I Word order
II NOM+V+OBL IIa
IIb
N/VInfl+S-V Conc
IIa detto +OBL Refl
III ABS+V+OBL III N/VInfl+S-V-O Conc
B NAERG – SP/FOCABS
Ergative Non-Agent
combined with Absolutive
Specifier/Focus
IX ERG+V+ABS IXa
IXb
N/VInfl+S-V-O Conc
V Infl +S-V-O Conc
C NAOBL–SP/FOCNOM/ABS
ObliqueNon-Agent
combined with Nominative
Specifier/Focus
IV NOM+V+OBL
IV ABS+V+OBL
IV ABS+V+OBL
IVa
IVb
IVc
N/VInfl+S-V Conc
N/VInfl+S-V-O Conc
VInfl+S-V-O Conc
V NOM+V+OBL V N/VInfl+ REFL
VI NOM+V+PREPOBL VI Prep + nominal form
D NAPOSS in SPNOM
NAPOSS in SPOBL
Possessive Non-Agent in
Nominative/Oblique
Specifier
VII NOMPOSS+V
VII NOMPOSS+V
VII NOM+V+OBLPOSS
VII
NOM+V+PREPOBLPOSS
VIIa
VIIb
VIIc
VIId
Determiner
Nominal inflection
Determiner
Determiner + Prep
E NAGEN in SPNOM
Genitive Non-Agent in
Nominative Specifier
VIII NOMGEN+REFLV VIII Genitive postmodifier
82
Table 5: Distribution of CA-frames in the sample, along with the genetic and morho-syntactic
typology
Genetic
type
Morpho-
syntactic
type
Cognitive
chain NON-AGENT + FOCUS/SPECIFIER Preva
iling
CA-
frame
CA
frame
A
B
C D
E
Coding
type
I II III IX IV V V
I
VII VI
II
Indo-
Europe
An
Isolating END 6 (1) 1 3 A
Isolating DAN 7 1(1) 2 (1) A
Isolating SW 7 1(1) 2 (1) A
Isolating IT 2(1) 5(3) 1 2 C
Isolating SP 2 5 2 1 C
Isolating PG 5
(1)
3 1 A
Inflect. SK 4(1) 6 (3) (1) C
Inflect. LITH 3 7(1) C
Inflect.
BG 5(1) 3 1 1 1 A
Ugro-
Finnic
Agglut. HU 1 4(2) 1 1 2 1 A
Basque Agglut. BASQ 1 5 2(1) 1 B
Cauca-
Sian
Agglut.
GEO 2(1) 1 (1) 8 C
Bracketed numbers show second-choice alternatives of respondents, however, they were
counted as relevant for the overall CA-frame characteristics.
As for the coding markers, word order was used as the sole activator of the NANOM/ABS –
SP/FOCOBL contrast in the prototypical analytical languages, namely English, Danish and
Swedish. These languages were quite consistent in coding their Non-Agents by word order with
minor exceptions of the Possessive form of pronouns as activators of the relationship between
the Perceiver and its Specifier. The Argument polysemy of the NOM+OBL surface alignment
frame in terms of its being used for both intentional and unintentional Argument chains was
compensated by the employment of either causative semi-auxiliaries and distinct lexical verbs,
or both (see sentences (8) and (12) in Appendix and also the commentary in Chapter 3.4). From among the rest of the languages in our sample, the NANOM/ABS – SP/FOCOBL CA-
frame also prevailed in Hungarian: the Non-Agent was coded as NOM in 7 out of 10 sentences,
in HU, in Bulgarian and Portuguese: 6 out of 10 sentences (quite interestingly, in Slovak: 5 out
of 10 sentences were classified as A/II). Thus, although English, Danish, Swedish, Portuguese
and Bulgarian, on the one hand, and Hungarian, on the other, are ranked with the distinct genetic
and morphological types (the former being analytical/inflective Indo-European languages, the
latter is a synthetic/agglutinative Ugro-Finnic language), when it comes to the activation of
Non-Agent, they employ the same Cognitive Alignment Frame (although Hungarian as an
agglutinative synthetic language activates the said contrast through inflectional endings rather
than by word order). The rest of the languages, whether Romance, Slavonic, Basque or
Georgian, showed preference for the NAOBL–SP/FOCNOM/ABS CA-frame.
83
As data in Table 5 indicate, distribution of CA-frames does not thoroughly correspond
to the genetic and morho-syntactic classifications. The major CA-frames for the Indo-European
languages and the only Ugro-Finnic language in the sample were A and C, however, distributed
unevenly across the genera (the only consistent genus in the sample seems to be the Germanic
languages). Basque and Georgian were identified to employ all the three major CA-frames,
with B being preferred in Basque and C in Georgian.
As for morpho-syntactic typology, analytical/isolating languages (6 in our sample) used
both the A and C CA-frames, with A prevailing in four of them (EN, DAN, SW, PG), two of
them (IT and SP) prefer the C CA-frame, however, using oblique nominal forms of pronouns
as the coding marker (IVa coding marker) rather than the diagnostic analytical marker, i.e. word
order. As for the three agglutinative languages, Hungarian, Basque and Georgian, each has its
preferred CA-frame, i.e. A, B, and C, respectively. In the inflectional/fusional languages in our
sample, the C CA-frame seems to prevail except for Bulgarian which prefers the A CA-frame.
Quite interestingly, Basque and Georgian do not employ the ERG case analogically,
with Basque as if being more ERG-oriented. Another observation of a difference between
Basque and Georgian is that in Georgian there were identified several instances of case
syncretism (cf. sentences (7) and (8)) where the overt pluripersonal concord was accelerated in
that the nominal form (pronoun me) was activated as DAT (7) or as ERG (8) solely by the
verbal inflection:
7 GEO Me k’alami
damivarda.
‘I a pen dropped.’
unwilled action
OSV me PRON.1.DAT.SG
PLRPC
m- prefix of V
da-m-i-vard-a V
PST-1.DAT.SG-
VERS-vard-
3.ABS.SG
k’alami
N.ABS.SG
IVb/C PLRPC
-m-DAT Object
-a-ABS
Subject
8 GEO Me davagde k’alami.
‘I dropped a pen.’
willed action
SVO me PRON.1.ERG.SG
PLPRC
-v- ERG INFL INFIX
of V
da-v-a-gd-e V
PST-1.ERG.SG-
VERS-gde-ABS
k’alami
N.ABS.SG
IXb/B
-v-ERG infix
(activating
willed action)
In Basque no such case was identified. Another interesting observation was the existence of the
coding alignment subtype IXb (covert ERG Subject activated by the verb´s inflection solely)
where the Subject valency element in ditransitive surface alignment was not expressed overtly
but only activated by the inflection of the verb:
9GEOii pexi movit’exe
‘I broke my
leg.’
(O)SV (me) PRON.1.ERG.SG
activated by -v- infix of V
under PLRPC
mo-v-i.t’ex-e
PST-1.ERG.SG-
VERS-break-ABS
pexi N.ABS.SG
And similar example in Basque for IVc coding alignment were the verb´s inflection activated
the Non-Agent as the sole marker in sentences (1) and (5):
5
BASQ
John pozik
dagoela iruditzen
zait.
SclV Covert 1.DAT.SG
activated by
-i-DAT INFL
INFIX of
DAT/ABS AUX
za-i-t under
PLRPC
iruditzen-V
za-i-t-DAT/ABS
AUX
za-1.DAT.SG-
3.ABS.SG
John
pozik
dagoela
(SUBO
RD,
FINITE
CLAUS
E)
IVc/C PLRPC
-i-.DAT Object
-t ABS Subject
84
3.4 What were the major coding markers and CA-frames for intentional argument chains?
Respondents were also invited to supply two intentional Argument chains, namely Intentional
Performer+Theme in sentence (8) (I dropped a pen) and Permitter+Resultant in sentence (12)
(John grows a beard in winter). It was found that the AGNOM/ABS – PAOBL CA-frame prevailed
consistently for both sentences (8) and (12), except for Basque having the AGERG – PAABS CA-
frame in both sentences, and Georgian having the AGERG–PABS CA-frame in sentence (8),
(probably because of the Aorist), and the AGNOM/ABS – PAOBL CA-frame in (12) (probably
because of the Present), along with the employment of a distinct verb base for the willed and
unwilled action.
Comparing sentences (7) and (8), only English and Hungarian (also Italian in (9ii))
displayed alignment polysemy of the same formal realization of both the unintentional (7) and
intentional (8) variants of I dropped a pen. The only indicator admitted by the Hungarian
respondent to distinguish between (7) and (8) was to insert ‘szándékosan/deliberately’ in
sentence (8) (the same as in English, actually), otherwise sentence (8) was ambiguous due to
its formal surface identity with (7). All the other languages, quite consistently, used a different
lexical base to distinguish the intentionality of the action. Some of them (SW, DEN, IT, PG)
also added a causative pseudomodal auxiliary, and, with Permitter, also a reflexive particle was
added (IT, SP, SK, LITH, BG).
Contrasting intentional and unintentional Argument chains in GEO and BASQ, Ergative
surface alignment was used for both intentional and Non-Agent cognitive chains. While in
BASQ the intentional variants were rendered in ERG+V+ABS surface alignment, in GEO the
latter was used only in the Aorist in (8). However, since the ERG surface alignment was
distributed across both the intentional and Non-Agent cognitive chains, its choice seemed to be
motivated grammatically, i.e. by surface valency ties + tense of the verb in GEO. On the other
hand, the GEO respondent admitted sentence (9) (I broke my leg) in ERG+V+ABS, while
sentence (10) (The car broke its axle.) was inadmissible in ERG+V+ABS with her reasoning
that in ERG sentence (10) would sound as if the car did it on purpose. Also if sentences (7) and
(8) were contrasted as to intentionality (see Chapter 3.3), in the willed variant the Agent was
coded as ERG, whereas in the unwilled variant the Patient received the ABS case although both
verbs were in the past tense. This somehow indicates that it need not be solely the surface ties
which motivate the choice of ERG+V+ABS surface alignment, but that there might be some
cognitive motives as well. This would, however, require further examination of the
phenomenon.
3.5 Testing the applicability of the Haspelmath’s EXPNOM SAE feature for the languages in the
sample
One of the 12 SAE features postulated by Haspelmath (2001), namely “a preponderance of
generalizing predicates to encode experiencers” (2001:1492–1510), assumes that Experiencers
appear as surface Subjects in NOM case in the SAE languages. From among the five cognitive
alignment frames outlined above, it is the A CA-frame NANOM/ABS – SP/FOCOBL which meets
the parameters of this assumption (coding alignments I, II, III). Although in our research the
category of Non-Agent arguments included beside Experiencers also Unintentional Performer
and Permitter, the results might be taken as showing some trends.
85
As can be derived from Table 5, 6 out 12 SAE languages in our sample show preference
for personal surface expression of Non-Agent, employing either analytical or inflectional
coding markers for that purpose. 5 other languages prefer the C CA-frame, and 1 language
employs B CA-frame. Although the number of languages is negligible for any SAE-relevant
conclusions, our analysis, at least, indicates that an im/personal way of expressing Non-Agents
is not bound to any of the morpho-syntactic or genetic types exclusively, and that each of the
languages sampled was able to activate its Non-Agents via both A and C CA-frames.
On the other hand, depending on a particular category of Non-Agent there may be
observed a kind of prioritization of certain CA-frames, for example, the A (personal) CA-frame
was identified in 10 out of 12 languages for sentence (4) I-Perceiver saw the man., or in 8 out
12 languages for sentence (9) I-Unintentional Performer broke my leg. On the other hand, in
the sub-category of Cognizer the C (impersonal) CA-frame was preferred in 11 out of 12
languages for sentence (5) It seems to me that John….
As was mentioned above, the C CA-frame diachronically preceded the A CA-frame in
the Indo-European language family, so it might be quite legitimate to hypothesize that the A
CA-frame used to convey impersonal cognitive chains in the SAE area is due to language
contact, as it was identified across genetic and morho-syntactic types of the languages in our
sample. However, these preliminary findings do not indicate its massive synchronic prevalence
(even the group of analytical languages does not seem to be uniform in this respect), and further
research involving the SAE-relevant number of languages is required to either prove or disprove
this eurouniversal assumption.
4. Conclusions
The research summarized in this paper showed that an onomasiological stance may be quite
successfully applied in syntactic typology since it allows of examining language-specific
combinations of coding markers employed to activate the cross-linguistically shared cognitive
argument chains. It actually proved that each of the respondents in the languages sampled was
capable of identifying the respective cognitive chains and rendering them by using their own
language-specific flagging. In our sample ten types of coding alignments were identified for
the selected cognitive chains.
The surface arrangements of coding markers rendered five Cognitive Alignment
Frames, namely NANOM/ABS – SP/FOCOBL, NAERG – SP/FOCABS, NAOBL– SP/FOCNOM/ABS,
NAPOSS in SPNOM/NAPOSS in SPOBL, and NAGEN in SPNOM. Each of the languages in the sample
employed at least two of the five CA-frames, favouring one of them. We identified a certain
language-specific preference for the respective CA-frames that did not thoroughly overlap with
the genetic and morphosyntactic types of languages. Drawing on the genetic and morho-
syntactic typology, this cognitive typological approach revealed that the genetically and
morphologically distant languages may, in fact, prefer the same CA-frames, for example:
English /Analytical-Germanic-Indo-European, Hungarian /Agglutinative Ugro-Finnic,
Bulgarian/Analytical-Slavic-Indo-European all prefer the NANOM/ABS – SP/FOCOBL CA-frame,
while the agglutinative ergative languages BASQ and GEO proved to prefer the NAERG –
SP/FOCABS and SP/FOCNOM/ABS - NAOBL, respectively.
The NOM/Subject Experiencer did not prove as the prevailing, first-choice, surface
alignment in all of the SAE languages in our sample. Nevertheless, the results of this study are
indicative only in this respect, limited by the goals of the research and the size of the corpus,
86
and further extensive examination would be necessary to draw linguistically substantiated
conclusions.
Abbreviations:
ABS Absolutive
ACC Accusative
ADJ Adjective
ADV Adverb
AG Agent
ART Article
AUX Auxiliary
BASQ Basque
BG Bulgarian
CONC Concord
CAUS Causative pseudomodal auxiliary
CA Cognitive alignment
DAN Danish
DAT Dative
EN English
ERG Ergative
EXP Experiencer
F Feminine
FOC Focus
GEO Georgian
HU Hungarian
INF Infinitive
INFL Inflectional
INS Instrumental
IT Italian
LEX distict lexical base
LITH Lithuanian
LOC Local
M Masculine
N Noun
NA Non-Agent
NOM Nominative
OBL Oblique
PA Patient
PG Portuguese
PL Plural
PLRPC Pluripersonal concord
PREP Preposition
PRON Pronoun
PRS Present
PST Past
PTCP Participle
87
REFL Reflexive
SG Singular
SP Specifier
SK Slovak
SP Spanish
SW Swedish
TH Theme
V Verb
Appendix Analysis of sentences
Analysis of Sentence 1 I like that picture.
Lg Sentence WO Emoter Emotive state Emotion focus Coding
marker/
Alignm
ent type
EN I like that picture. SVO I PRON.1.NOM.SG like-0 V.PRS-1.SG that PRON
picture N.
OBL.SG
I/A
DAN Jeg synes om det billed. SVO jeg PRON.1.NOM.SG synes-0 V
like-1.SG
om-PART
det ART
billed N.
OBL.SG
I/A
SW Jag
tycker om den bilden.
SVO jag PRON.1.NOM.SG tycker-0 V
like-1.SG
om-PART
den ART
bilden N.
OBL.SG
I/A
IT Mi piace quella foto. OVS mi PRON.1.DAT.SG
-e-V Infl.3.SG
piac-e V
like-3.SG
quella PRON
foto N.
NOM.SG
IVa/C
SP Me gusta esta foto. OVS me PRON.1.DAT.SG
-a-V Infl.3.SG
gust-a V
like-3.SG
esta PRON
foto N.
NOM.SG
IVa/C
PG Eu gosto deste quadro.
SVO eu PRON.1.NOM.SG
-o-V INFL.1.SG
gost-o V
like-1.SG
deste
PREP+PRON
quadro
N.OBL.SG
IIa/A
SKi 1.Páči sa mi ten obraz.
VOS mi PRON.1.DAT.SG páči-0 V
like-3.SG
sa-REFL PART
ten PRON
obraz
N.NOM.SG
IVa/C
SKii 2.Mám rada ten obraz. (S)VO (ja)PRON.1.NOM.SG
activated by-m- V
INFL.1.SG
má-m AUX
have-1.SG
rad-a-ADJ
glad-F
ten PRON
obraz
N.ACC.SG
IIa/A
LITH Man patinka tas
paveikslas.
OVS man
PRON.1.DAT.SG
patinka V.3.SG/PL
tas PRON
paveikslas
N.NOM.SG
IVa/C
BGi Тази картина ми
харесва.
SOV mi PRON.1.DAT.SG
-a-V INFL.3.SG
haresv-a V
like-3.SG
tazi PRON
kartina
N.NOM.SG
IVa/C
BGii Аз харесвам тази
картина.
SVO az PRON.1.NOM.SG
-am-V INFL.1.SG
haresv-am V
like-1.SG
tazi PRON
kartina
N.OBL.SG
IIa/A
HUi Tetszik nekem az a kép.
‘That picture pleases for
me.’
VOS nekem-
PREP+PRON.1.OBL.
SG
‘for me’
tetszik-0 V
like-3.SG/INF
az
PRON.ACC.SG
a-ART
kép N.NOM.SG
VI/C
88
HUii Kedvelem azt a képet.
‘I like that picture.’
(S)VO (én)
PRON.1.NOM.SG
activated by
-em-V INFL.1.SG
kedvel-em V
like-1.SG/INF
azt PRON
a ART
kép-et N
kép-ACC.SG
IIa/A
BASQ
i
Nik hori atsegin dut.
‘I like that.’
SOV ni-k PRON
ni-1.ERG.SG
-t ERG INFL SUFFIX
of V under PLRPC
atsegin-V
d-u-t-ERG/ABS AUX
ABS.3.SG-u-ERG.SG
‘I have it’
hori PRON.
3.ABS.SG
IXa/B
PLRPC
-t ERG
Subject
-d ABS
Object
BASQ
ii
Hori atsegin zait.
‘That pleases me.’
SV Covert 3.DAT.SG
activated by
-i- DAT INFL INFIX
of DAT/ABS AUX
za-i-t under PLRPC
atsegin-V
za-i-t-DAT/ABS
AUX
‘it/s/he is to me’
za-1.DAT.SG-
3.ABS.SG
hori PRON.
3.ABS.SG
IVc/C
PLRPC
-i-DAT
Object
-t ABS
Subject
GEO Me momc’ons is surati.
‘That picture pleases
me.’
OVS me PRON.1.DAT.SG
activated by -m- DAT
INFL INFIX of V
under PLRPC
mo-m-c’on-s V
PRV-DAT.SG-like-
3.ABS.SG
is PRON
surati N.
ABS.SG
IVc/C
PLRPC-
m-DAT
Object
-s ABS
Subject
Commentary:
A Danish/Swedish respondent admitted that historically they had the form mig synes … ‘(to) me {OBL; V-passive}, also
pointing out that the verb conjugation in Danish and Swedish has no person differentiation, and that there is no case inflection
in nouns, except for the -s Possessive case.
In BASQi dut is the bivalent transitive (ABS/ERG) present and past indicative form of ´edun/ezan´ to have´ for a 3rd person
ABS argument: -t manifests concord with the first person singular Ergative Experiencer Subject. At the same time the d-
manifests concord with the third person Absolutive Focus Object (Hualde 2003: 210, 221-222).
In BASQii zait is the bivalent intransitive (ABS/DAT) present indicative form of izan ´to be´ for a 3rd person ABS argument:
-i- infix indicates Dative Experiencer – ‘to me’, while -t manifests concord with ABS Stimulus. (Hualde 2003: 210, 212-213).
Basque as a highly inflectional language encodes arguments within the internal morphological structure of the verb:
dakar-ki-zu-t (bring-it-to you-I) ki-Theme, zu-Recipient, t-Agent
dakar-ki-da-zu (bring-it-to me-you) ki-Theme, da-Recipient, zu-Agent (Hualde 2003: 209)
In BASQi the Experiencer is the Ergative Subject and the Focus/Stimulus is the Absolutive Object. In BASQi The
Focus/Stimulus is the Absolutive Object, and the Experiencer is the Dative Subject. The alternance is reminiscent of the
difference between the Latin temere class (temere/fear class with Subject Experiencer) and the piacere class (piacere/appeal to
class taking object Experiencer). (Etxepare 2003:44-45)
In Basque two surface strings are available:
SOV: ERG/S+ABS/O + V+ABS/ERG AUX (dut) = ‘I that like’, where the auxiliary manifests concord with the ERG
Experiencer in the inflectional suffix -t, or
SV: ABS/S+V+ABS/DAT AUX (zait) = ‘That pleases me’, where the auxiliary manifests concord with the ABS
Focus/Stimulus in the inflectional suffix -t-, and, moreover, special attention deserves the fact that the infix -i- activates DAT
Experiencer, as a sole marker.
Hence, there is a choice between the transitive and intransitive auxiliaries (dut/ zait) and inflectional items indicating the
syntactic arguments. As can be seen in one string Subject Experiencer is ERG-marked and the required concord is displayed in
the selected ERG/ABS auxiliary. Whereas in the other string the Experiencer is coded in the ABS/DAT auxiliary.
In GEO me is a case-neutral form of personal pronoun, its ERG or DAT reading is activated by the inflection of the verb
based on the rules of pluripersonal concord.
As a generalizing observation it can be pointed out that NOM arguments, whether fixed by word order or nominal/verbal
inflection, occur in SVO word order.
89
Analysis of Sentence 2 I feel sorry.
Lg Sentence WO Emoter Emotive state Emotion focus Coding
marker
/Alignment
type
EN I feel sorry. SVA I PRON.1.NOM.SG feel-0 V.PRS-
1.SG
sorry ADJ/ADV
Context based
(about that)
I/A
DAN Jeg er ked af det. SVA jeg PRON.1.NOM.SG er
AUX/copula.1.SG
ked-ADJ
af det PREP+
PRON
I/A
SW Jag är ledsen. SVA jag PRON.1.NOM.SG är
AUX/copula.1.SG
ledsen-ADJ
Context based I/A
IT Mi dispiace. (S)OiV mi PRON.1.DAT.SG
-e-V INFL3.SG
dispiac-e V
feelsorry-3.SG
Contex-based IVa/C
SP Lo siento. (S)OV -o V INFL.1.SG sient-o V
feelsorry-1.SG
Lo PRON.
3.OBL.SG
IIa/A
PGi Sinto-me
arrependido/a.
(S)VOA (yo) PRON.1.NOM.SG
activated by
-o-V INFL.1.SG
me PRON.1.ACC.SG
sent/sint-o V
feel-1.SG
arrependid-o/a
sorry-SG.M/F
context based X/A-C
PGii Tenho pena.
‘I feel regret.’
(S)VO (yo) PRON.1.NOM.SG
activated by
-o-V INFL.1.SG
tenh-o V
feel-1.SG
pena-N
regret-ACC.SG
context based IIa/A
SKi Je mi (to) ľúto.
‘Is to me (that)
sorry.’
VO(S)A mi PRON.1.DAT.SG je AUX
be.3.SG
ľúto ADV
sorry ADV
to PRON.3.
NOM.SG
IVa/C
SKii Mrzí ma to.
‘It feels sorry to
me.’
VOS ma PRON.1.ACC.SG mrz-í V
feelsorry-3.SG
To PRON.3.
NOM.SG
IVa/C
LITH Man gaila.
O(V)A man PRON.1.DAT.SG (yra) V PRS.3.SG
copula ‘be’ can be
omitted in the
present tense
gaila ADV IVa/C
BG Съжалявам.
(S)V (az) PRON.1.NOM.SG
activated by
-m-V INFL.1.SG
sâzhalyava-m V
feelsorry-1.SG
Contex-based IIa/A
HU Sajnálom. (S)V (én) PRON.1.NOM.SG
activated by
-om-V INFL.1.SG
sajnál-om V
feelsorry-1.SG
Contex-based IIa/A
BASQ Barkatu.
(S)V unexpressed emoter barkatu V
impersonal
Contex-based Impersonal
verb form
GEO (Me) vc’uxvar. (S)V (me) PRON.1.ABS.SG
activated by v- INFL
PREFIX of V under
PLRPC
v-c’ux-var V
ABS.S-sorry-be
Contex-based
III/A
v-ABS
Subject
Commentary:
PGi is interesting in combing NOM nominal and verbal inflection, however, the 1SG verb takes a pronominal particle -me that
was interpreted by the respondent as the Accusative case. So it seems as a hybrid subtype in between II and IV types.
90
Analysis of Sentence 3 My leg hurts.
Lg Sentence WO Perceiver-
whole
Perception state Perceiver-part Coding
marker/
Alignment type
EN My leg hurts. SV my PRON
POSS.1.SG
hurt-s V.PRS-
3.SG
leg N.NOM.SG VIIa/D
DANi Mit ben gør ondt.
‘My leg does bad.’
SVA mit PRON
POSS.1.SG
gør V.3.SG&PL
ondt ADJ
ben N.NOM.
SG&PL
VIIa/D
DANii Det gør ondt i mit
ben.
‘It does bad in my
leg. ’
SVAA mit PRON
POSS.1.SG
gør V.3.SG
ondt ADJ
det formal Subject
i PREP
ben N.OBL.
SG&PL
VIIb/D
SWi Mitt ben gör ont. SVA mitt PRON
POSS.1.SG
gør V.3.SG&PL
ondt ADJ
ben N.NOM.
SG&PL
VIIa/D
SWii Det gör ont i mitt
ben.
SVAA mitt PRON
POSS.1.SG
gör V.3.SG
ont ADJ
det formal Subject
i PREP
ben N.OBL.
SG&PL
VIIb/D
ITi Mi fanno male le
gambe.
‘To me legs do
disease.’
OVS mi PRON.
1.DAT.SG
fanno V.3.PL
male N.M.SG
le ART
gambe N.NOM.PL
IVa/C
ITii Mi dolgono le gambe.
(formal register)
OVS mi PRON.
1.DAT.SG
dolgono V.3.PL le ART
gambe N.NOM. PL
IVa/C
SP Me duele la pierna.
OVS me PRON.
1.DAT.SG
duele V.3.SG la ART
pierna N.NOM.SG
IVa/C
PG Dói-me a perna. VOS me PRON.
1.DAT.SG
dói V.3.SG a ART
perna N.NOM.SG
IVa/C
SK Bolí ma noha. VOS ma PRON.
1.ACC.SG
bol-í V
hurt-3.SG
noha N.NOM.SG IVa/C
LITH Man skauda koją.
OVS man PRON.
1.DAT.SG
skauda V.3. SG koją N.NOM.SG IVa/C
BG Боли ме кракът. VOS me PRON.
1.ACC.SG
boli V.3.SG Krakât N.NOM.SG
+ART
IVa/C
HU Fáj a lábam.
VS -am-POSS.1.SG fáj V.3.SG a ART
láb-am N
NOM.POSS.1.SG
VIIc/D
BASQ Nik mina daukat nire
hankan.
SVO ni-k PRON
ni-1.ERG.SG
-t ERG INFL
SUFFIX of
AUX under
PLRPC
nire-POSS-1.SG
mina-V
d-auka-t
-ERG/ABS AUX;
ABS.3.SG-auka-
ERG.SG
hankan N.ABS.SG IXa/B
PLRPC
-t ERG Subject
-d ABS Object
GEO Pexi mt’k’iva.
(O)SV (me)
PRON.1.DAT.S
G activated by
m- INFL
PREFIX of V
under PLRPC
m-t’k’iv-a V
1.DAT.SG-
t’k’iv-3.ABS.SG
pexi N.ABS.SG IVc/C
m-DAT Object
-a-ABS Subject
Commentary:
In HU the Possessive case is activated synthetically by the nominal inflection rather than by a separate pronoun.
In BASQ the auxiliary daukat is a 1st person singular present indicative form of eduki ‘to have’ (transitive verbs) of a 3rd person
singular object; -t manifests pluripersonal agreement of the auxiliary with the Ergative Subject and d- with the Absolutive
Object. (Hualde 2003: 234). Moreover, the Possessive case is overtly realized by the possessive pronoun nire.
In GEO m-inflectional prefix of the verb activates the DAT Object Perceiver leaving it fully covert. This is possible based on
the pluripersonal concord rules (the combination of the m- implied Object Perceiver and -a implied ABS Subject/Part of
Perceiver (pexi/ABS), with no possessive indicator in contrast to BASQ.
91
Analysis of Sentence 4 I saw the man.
Lg Sentence WO Perceiver Perception State Perception focus Coding
marker
/Alignme
nt type
EN I saw the man.
SV
O
I PRON.1.NOM.SG saw-0 V.PST-1.SG the ART
man N.OBL.SG
I/A
DAN Jeg så manden. SV
O
jeg PRON.1.NOM.SG så-0 V
saw-1.SG
mand-en N
OBL.SG-ART
I/A
SW Jag såg mannen. SV
O
jag PRON.1.NOM.SG såg-0 V
saw-1.SG
mann-en N
OBL.SG-ART
I/A
IT Ho visto l’uomo.
(S)V
O
(io) PRON.1.NOM.SG
activated by
-o-V INFL.1.SG
h-o-AUX
have-1.SG
vist-o-PST PTCP
see-M.SG
l’-uomo
ART-N.OBL.SG
IIa/A
SP Vi al hombre.
(S)V
O
(yo) PRON.1.NOM.SG
activated by
-i-V INFL.1.SG
v-i V
see-PST.1.SG
al ART
hombre N.OBL.SG
IIa/A
PG Vi o homem.
(S)V
O
(eu) PRON.1.NOM.SG
activated by
-i-V INFL.1.SG
v-i V
see-PST.1.SG
o ART
homem N.OBL.SG
IIa/A
SK Videl som toho
muža.
(S)V
O
(ja) PRON.1.NOM.SG
activated by
som-AUX
be-PRS.1.SG
vide-l-0 V
see-PST-M
som-AUX
be-PRS.1.SG
toho PRON
muža N.
ACC/OBL.SG
IIa/A
LITH Pamačiau žmogų. (S)V
O
(aš) PRON.1.NOM.SG
activated by V INFL
pamačiau V.1;SG žmogų N.
OBL/ACC.SG
IIa/A
BG (Аз)Видях мъжа.
(S)V
O
(az) PRON.1.NOM.SG
activated by
-h-V INFL.1.SG
vidya-h V
see-PST.1.SG
mâzha N.
OBL/ACC.SG
IIa/A
HU Láttam az embert. (S)V
O
(én) PRON.1.NOM.SG
activated by
-m-V INFL.1.SG
lát-ta-m V
lát-PST-1.SG
az ART
embert N.
OBL/ACC.SG
IIa/A
BASQ Nik gizona ikusi
nuen.
SO
V
ni-k PRON
ni-1.ERG.SG
n- ERG INFL PREFIX of
AUX under PLRPC
ikusi-V
n-u-en ERG/ABS
AUX
ERG.SG-u-
ABS.3.SG
gizona N.ABS.SG IXa/B
PLRPS
n- ERG
Subject
-en ABS
Object
GEO Me davinaxe
k’aci.
SV
O
me PRON.1.ERG.SG
PLPRC
-v- ERG INFL INFIX of
V under PLRPC
da-v-i-nax-e V
PST-ERG.SG-
VERS-nax-ABS.SG
k’aci N.ABS.SG IXb/B
PLRPS
-v- ERG
Subject
-e- ABS
Object
Commentary:
In BASQ the bivalent transitive ABS/ERG past indicative form of ´edun/ezan´ to have´ is used to activate a third person ABS
argument in combination with the ERG Subject. In the present tense the ERG Argument (the Subject) is encoded by a suffix
(du-t) (see the analysis of sentence 1 in BASQi), in the past by a prefix n-. (Hualde 2003:222). It is so called ergative
displacement (ibid 207).
92
Analysis of Sentence 5 It seems to me that John is happy.
Lg Sentence WO Cognizer Cognition state Cognition
Focus
Coding
marker
/Alignme
nt type
EN It seems to me
that John is
happy.
SVOprepScl to PREP
me PRON.
OBL.SG
seem-s V.PRS-3.SG it formal Subject
that John is happy
(SUBORD.
FINITE
CLAUSE)
VI/C
DAN Det forekommer
mig at John er
glad.
SVOiScl mig
PRON.1.OBL.SG
forekommer-0 V
seem-3.SG
det formal Subject
at John er glad
(SUBORD
FINITE
CLAUSE)
IVa/C
Word
Order
rather
than SV
concord is
employed
SW För mig det
verkar som att
John är glad.
OprepSVScl för PREP
mig PRON.1.
OBL.SG
verkar-0 V
seem-3.SG
det formal Subject
som att John är
glad (SUBORD
FINITE
CLAUSE)
VI/C
ITi Mi sembra che
John sia felice.
OVScl mi PRON.1.
OBL/DAT. SG
sembr-a V
seem-3.SG
che J. sia felice
(SUBORD.
FINITE CLAUSE
in SUBJUNC
TIVE MOOD)
IVa/C
ITii John mi sembra
felice.
SOVA mi PRON.1.
OBL/DAT. SG
sembr-a V
seem-3.SG
John N.NOM.SG
felice ADJ- split
cognitive unit
IVa/C
SP Me parece que
John esta
contento.
OVScl me PRON.1.
OBL/DAT.SG
parece V.3.SG que John esta
contento
(SUBORD
FINITE
CLAUSE)
IVa/C
PG Parece-me que o
João está feliz.
VOScl me PRON.1.
OBL/DAT.SG
parece V.3.SG. que o João está
feliz (SUBORD.
FINITE
CLAUSE)
IVa/C
SKi Zdá sa mi, že
John je šťastný.
VOScl mi PRON.1;
OBL/DAT.SG
zdá-0 V
seem-3.SG
sa-REFL PART
že John je šťastný
(SUBORD
FINITE
CLAUSE)
IVa/C
SKii John sa mi zdá
šťastný.
SOVA mi PRON.1.
OBL/DAT.SG
zdá V.3.SG
sa REFL PART
John N.NOM.SG
šťastný ADJ
split cognitive unit
IVa/C
LITH Man atrodo, kad
Jonas laimingas.
OVScl man PRON.1.
OBL/DAT.SG
atrodo V.3.SG kad Jonas
laimingas
(SUBORD.
FINITE
CLAUSE)
IVa/C
BG Струва ми се,
че Джон е
щастлив.
VOScl mi PRON.1.
OBL/DAT.SG
struva V.3.SG
se REFL PART
che Dzhon e
tshastliv
(SUBORD.
FINITE
CLAUSE)
IVa/C
HU Úgy vélem,
(hogy) János
boldog.
AVScl
(én) PRON.1.
NOM.SG
activated by
-em-V INFL.1.SG
úgy ADVL
vél-em V
seem-1.SG
(hogy) János
boldog
(SUBORD.
FINITE
CLAUSE)
IIa/A
93
BASQ John pozik
dagoela iruditzen
zait.
SclV Covert 1.DAT.SG
activated by
-i-DAT INFL
INFIX of
DAT/ABS AUX
za-i-t under
PLRPC
iruditzen-V
za-i-t-DAT/ABS
AUX
za-1.DAT.SG-
3.ABS.SG
John pozik
dagoela
(SUBORD,
FINITE
CLAUSE)
IVc/C
PLRPC
-i-.DAT
Object
-t ABS
Subject
GEO Me mečveneba,
rom džoni
bednieria.
OVScl me PRON.1.
DAT.SG
PLRPC
m- DAT INFL
PREFIX of V
under PLRPC
m-e-čveneb-a V
1.DAT.SG-PASS-
čveneb-3.ABS.SG
rom džoni
bednieria.
(SUBORD.
FINITE
CLAUSE)
IVc/C
PLRPC
m-DAT
Object
-a-ABS
Subject
Commentary:
Similarly as in Emoter in sentence 1, BASQ activates DAT Cognizer solely by the verb´s inflection.
The only language activating Cognizer in this sentence by NOM is Hungarian. All the other languages employ OBL cases.
Analysis of Sentence 6 I am interested in music.
Lg Sentence WO Cognizer Cognition state Cognition
focus
Coding
Marker/
Alignme
nt type
ENi
I am interested in
music.
SVCsA I PRON.1.NOM.SG am-
AUX.PSR.1.SG
interested-PTCP
in PREP
music
N.OBL.SG
I/A
ENii Music interests me. SVO me PRON.1.OBL.SG interest-s V
interest-3.SG
music
N.NOM.SG
IVa/C
DANi
Jeg er interesseret i
musik.
SVCsA jeg PRON.1.NOM.SG er AUX
be.1.SG
interesseret -ADJ
i PREP
music
N.OBL.SG
I/A
DANii Musik interesserer mig. SVO mig PRON.1.OBL.SG interesserer-0
interest-3.SG
music
N.NOM.SG
IVa/C
SWi Jag är intresserad av
musik.
SVCsA jag PRON.1.NOM.SG är AUX
be.1.SG
intresserad –ADJ
av PREP
music
N.OBL.SG
I/A
SWii Musik intresserar mig. SVO mig PRON.1.OBL.SG intresserar-0 V
interest-3.SG
music
N.NOM.SG
IVa/C
ITi Sono interessato alla
musica.
(S)VCs
A (io) PRON.1.NOM.SG
activated by
-o-V INFL SUFFIX.1.SG
son-o AUX
be-M.1.SG
interesat-o-PST
PTCP
interested-M.1.SG
alla
PREP+ART.
F
musica N.
OBL.SG
IIa/A
ITii Mi interessa la musica. OVS mi PRON.1.OBL/DAT.SG interess-a V
interest-3.SG
la ART
musica N.
NOM.SG
IVa/C
SP Me interesa la musica. OVS me PRON.1.OBL/DAT.SG interes-a V
interest-3.SG
la ART
musica N.
NOM.SG
IVa/C
PG Estou interessado/a em
música.
(S)VCs
A (eu) PRON.1.NOM.SG
activated by
-ou-V INFL SUFFIX.1.SG)
est-ou-AUX
be-1.SG
interessad-o-ADJ
interested-1.M
em PREP
música N.
OBL.SG
IIa/A
SKi Zaujímam sa o hudbu. (S)VA (ja) PRON.1.NOM.SG
activated by
-m-V INFL SUFFIX.1;SG
zaujíma-m V
interest-1.SG
sa-REFL PART
o PREP
hudbu
N.LOC.SG
IIa/A
SKii Zaujíma ma hudba. VOS ma PRON.1.OBL/ACC.SG zaujím-a V
interst-3.SG
hudba
N.NOM.SG
IVa/C
94
LITHi Domiuosi muzika. (S)VA (aš) PRON.1.NOM.SG domiuosi V
interst-1.
SG.REFL
muzika
N.OBL/INS.
SG
IIa/A
LITHii Mane domina muzika. OVS mane PRON.1.
OBL/ACC.SG
domina V
interest-3.SG
muzika
N.NOM.SG
IVa/C
BG Интересувам се от
музика.
(S)VA (az) PRON.1.NOM.SG
-m-V INFL SUFFIX.1.SG
interesuva-m V
interest-1.SG
se-REFL PART
ot PREP
muzika
N.OBL.SG
IIa/A
HUi Érdekel engem a zene.
‘Music interests me.’
VOS engem PRON.1.OBL.SG érdekel V
interst-3.SG
a ART
zene
N.NOM.SG
IVa/C
HUii Foglalkozom a zenével.
‘I deal with music.’
(S)VA (én) PRON.1.NOM.SG
activated by
-om-V INFL SUFFIX.1.SG
foglalkoz-om V
deal-1.SG
a ART
zené-vel
N.OBL.SG
IIa/A
BASQ Ni musican interesatuta
nago.
SAV ni PRON.1.ABS.SG interesatuta V
nag-o AUX
nag-1.ABS.SG
musican N.
LOC.SG
III/A
OBL-
LOC
GEOi Me maint’eresebs
musik’a.
OVS me PRON.1.DAT.SG
PLRPC
m- prefix of V
m-a-intereseb-s V
1.DAT.SG-
VERS-interseb-
3.ABS.SG
musik’a
N.ABS.SG
IVc/C
PLRPC
m-DAT
Object
-s-ABS
Subject
GEOii Me daint’eresebuli var
musik’it.
SVA me- PRON.1.ABS.SG daint’eresebuli-
PTCP/ADJ
var-PASS.1.SG
musik’-it N
musik’-
OBL/INSTR.
SG
III/A
OBL-
INSTR
Commentary:
BASQ -an in musican is a locative inflectional ending (Hualde 2003:185)
In GEO, 1SG pronoun has the same form for both the DAT Object and ABS Subject (and also ERG Subject in sentence 4), so
it is the verb’s inflection that activates a particular DAT/ABS/ERG interpretation of the nominal element.
Analysis of Sentence 7 I dropped a pen. (unwilled action)
Lg Sentence WO Unintentional
Performer
Unwilled Action Specifier Coding
marker/
Alignme
nt type
EN I dropped a pen. SVO I PRON.1.NOM.SG dropp-ed-0 V
drop-PST-1.SG
a ART
pen N.OBL.SG
I/A
DAN Jeg tabte en pen. SVO jeg PRON.1.NOM.SG tabte V
drop-PST.1.SG
en ART
pen N.OBL.SG
I/A
SW Jag tappade en
penna.
SVO jag PRON.1.NOM.SG tappade V
drop-PST.1.SG
En ART
penna N.OBL.SG
I/A
IT Mi è caduta una
penna.
OVS mi PRON.1.
OBL/DAT.SG
é-0 AUX
be-3.SG
cadut-a
fall-PST PTCP.F.SG
una ART
penna N.
NOM.SG
IVa/C
SP Se me cayo el
boligrafo.
OVS me PRON.1.
OBL/.DAT.SG
se-REFL PART
cay-o V
fall-PAST.3.SG
el ART
boligrafo
N.NOM.SG
V/C
PG Caiu-me uma
caneta.
VOS me PRON.1.
OBL/.DAT.SG
caiu-0 V
fall-PST.3.SG
uma ART
caneta
N.NOM.SG
IVa/C
SK Spadlo mi pero. VOS mi PRON.1.
OBL/DAT.SG
spadl-o V
fall-PST.3.SG
pero N.NOM.SG IVa/C
LITH Man iškrito
pieštukas.
OVS man PRON.1.
OBL/DAT.SG
iškrito-0 V
fall-PST.3.SG
pieštukas
N.NOM.SG
IVa/C
BG Изпуснах един
химикал.
(S)VO (az) PRON.1.
NOM.SG activated by
izpusna-h V
drop-PST.1.SG
edin NUM
himikal
N.OBL.SG
IIa/A
95
-h-V INFL
SUFFIX.1.SG
HU Leejtettem egy
tollat.
(S)VO (én) PRON.1.
NOM.SG
-em (V Infl; 1;SG)
leejt-ett-em V
drop-PST-1.SG
egy ART
tollat
N.OBL/ACC.SG
IIa/A
BASQ Niri boligrafoa
erori zait.
OSV niri PRON.1.DAT.SG
-i- DAT INFL INFIX
of DAT/ABS AUX
za-i-t under PLRPC
erori V
za-i-t-DAT/ABS AUX
‘it/s/he is to me’
za-1.DAT.SG-3.ABS.SG
boligrafoa N.
ABS.SG
IVb/C
PLRPC
-i-DAT
Object
-t ABS
Subject
GEO Me k’alami
damivarda.
OSV me PRON.1.DAT.SG
PLRPC
m- prefix of V
da-m-i-vard-a V
PST-1.DAT.SG-VERS-
vard-3.ABS.SG
k’alami
N.ABS.SG
IVc/C
PLRPC
-m-DAT
Object
-a-ABS
Subject
Commentary:
The term Specifier is used to refer to a component of the Unintentional Perfomer, its belongings, body part etc.
Analysis of Sentence 8 I dropped a pen. (willed action)
Lg Sentence WO Intentional
Performer
Willed Action Theme Coding
Marker/Al
ignment
type
EN I dropped a pen
(intentionally).
SVO I PRON.1.NOM.SG dropp-ed-0 V
drop-PST-1.SG
a ART
pen NOBL.SG
I/A+ADVL
DAN Jeg lod en pen falde.
‘I let a pen fall.’
SVO jeg PRON.1.NOM.SG lod-0 AUX
let-1.SG
falde V INF
en ART
pen N.OBL.SG
I/A+
pseudomod
al causative
AUX+
lexical verb
SW Jag lät falla en penna.
‘I let a pen fall.’
SVO jag PRON.1.NOM.SG lät-0 AUX
let-1.SG
falla V INF
en ART
penna
N.OBL.SG
I/A+
pseudomod
al causative
AUX+
lexical verb
IT Ho fatto cadere una
penna.
‘I made a pen fall.’
(S)VO (io) PRON.1.
NOM.SG
activated by
-o-V INFL.1.SG
h-o AUX
have-1.SG
fatt-o-PST PTCP
made-M.SG
una ART
penna
N.OBL.SG
IIa/A+pseu
domodal
causative
AUX
SP He tirado el boligrafo.
‘I threw a pen.’
(S)VO (yo) PRON.1.
NOM.SG
activated by
-o-V INFL.1.SG
h-e AUX
be-1.SG
tirad-o PST PTCP
throw-M.SG
el ART
boligrafo
N.OBL.SG
IIa/A+lexic
al verb
PGi Atirei uma caneta.
‘I threw a pen.’
(S)VO (eu) PRON.1.
NOM.SG
activated by
-ei-V INFL.1.SG
atir-ei V
throw-PST.1.SG
uma ART
caneta N.OBL/
ACC.SG
IIa/A+lexic
al verb
PGii Eu deixei cair a caneta.
‘I let fall a pen.’
SVO eu PRON.1.NOM.SG
-ei-V INFL.1.SG
deix-ei AUX
let-PST.1.SG
cair V INF
a ART
caneta
N.OBL/ACC.S
G
IIa/A+
pseudomod
al causative
AUX
+ lexical
verb
SK
Z/odhodil som pero.
‘I threw a pen.’
(S)VO (ja)
PRON.1.NOM.SG
activated by
z/odhodi-l V
throw-PST.M.SG
som AUX
pero N.ACC.SG IIa/A+lexic
al verb
96
som-AUX
be-PRS.1.SG
be-PRS.1.SG
LITH Išmečiau pieštuką.
(S)VO (aš) PRON.1.
NOM.SG
išmečiau V
throw-PST.1.SG
pieštuką
N.ACC.SG
IIa/A+lexic
al verb
BG Хвърлих един
химикал.
(S)VO (az) PRON.1.
NOM.SG activated by
-h-V INFL
SUFFIX.1.SG
hvarli-h V
throw-PST.1.SG
edin NUM
himikal
N.OBL.SG
IIa/A+lexic
al verb
HU (Szándékosan)
Leejtettem egy tollat.
(A)(S)
VO
(én) PRON.1.
NOM.SG
-em-V INFL
SUFFIX.1.SG
leejt-ett-em V
drop-PST-1.SG
szándékosan
ADVL
intentionally
egy ART
tollat
N.ACC.SG
IIa/A+AD
VL
BASQ Nik boligrafoa bota
dut.
SOV ni-k PRON
ni-1.ERG.SG
-t ERG INFL SUFFIX
of V under PLRPC
bota V
d-u-t-ERG/ABS
AUX
3.ABS.SG-u-
ERG.SG ‘I have
it’
boligrafoa
N.ABS.SG
IXa/B
PLRPC
-t ERG
Subject
-d ABS
Object
GEO Me davagde k’alami.
SVO me PRON.1.ERG.SG
PLPRC
-v- ERG INFL INFIX
of V
da-v-a-gd-e V
PST-1.ERG.SG-
VERS-gd-ABS
k’alami
N.ABS.SG
IXb/B
v-ERG
infix
(activating
willed
action)
Analysis of Sentence 9 I broke my leg (unwilled action)
Lg Sentence WO Unintentional Performer Unwilled Action Specifier Coding
Marker/Al
ignment
type
EN I broke my leg.
SVO I PRON.1.NOM.SG
my-POSS.1.SG in OBL NP
broke-0 V.PST-
1.SG
leg N.OBL.SG VIId/D
DAN Jeg brækkede
benet.
SVO jeg PRON.1.NOM.SG brækkede V
break-PST.1.SG
benet
N+ART.OBL.S
G
I/A
SW Jag bröt benet. SVO jag PRON.1.NOM.SG bröt V
break-PST.1.SG
benet
N+ART.OBL.S
G
I/A
ITi Mi si è rotta una
gamba.
OVS mi PRON.1.DAT.SG
é AUX
be-3.SG
rott-a
break-PST
PTCP.F. SG
si REFL PART
la ART
gamba
N.NOM.SG
V/C
ITii Mi sono rotto la
gamba.
OVS (io) PRON.1.NOM.SG
activated by
-o V inflection
mi PRON.1.DAT.SG
son-o AUX
be-1.SG
rott-o
break-PST
PTCP.1.
SG
la ART
gamba
N.OBL.SG
IIb/A
SP Me he roto la
pierna.
OVS me PRON.1.DAT.SG h-e AUX
be-3.SG
rot-o
break-PST
PTCP.M.SG
la ART
pierna
N.NOM.SG
IVa/C
97
PG Parti uma
perna.
(S)VO (eu) PRON.1.NOM.SG
activated by
-i-V INFL.1.SG
part-i V
break-PST.1.SG
uma ART
perna
N.ACC.SG
IIa/A
SK Zlomil som si
nohu.
‘I broke to me a
leg.’
(S)VO (ja) PRON.1.NOM.SG
activated by
som-AUX
be-PRS.1.SG
si-REFL.DAT.SG
zlomi-l V
break-PST.M.SG
som AUX
be-PRES.1.SG
si-
REFL.1.DAT.SG
nohu
N.ACC.SG
IIb/A
LITH Susilaužiau
koją.
(S)VO (aš) PRON.1.NOM.SG susilaužiau V
REFL.1.SG
koją N.ACC.SG IIb/A
BG Счупих си
крака.
(S)VO (az) PRON.1.NOM.SG
activated by
-h-V INFL SUFFIX.1.SG
si-PRON REFL.1.
OBL/DAT.SG
schupi-h V
break-PST.1.SG
si-PRON
REFL.1.DAT.SG
krakâ
N.OBL.SG+AR
T
IIb/A
HU Eltört a lábam.
‘I broke my
leg.’
VS -am NOM.POSS.1.SG eltört V
break-PST.3.SG
a ART
láb-am N.
NOM.POSS.1.S
G
VIIc/D
BASQ Nire hanka
apurtu dut.
SV nire- PRON POSS.1.SG apurtu-V.PERF
dut-AUX.ABS
(present perfect
consists of perfect
verb form +
indicative form of
AUX –
intransitive)
hanka
N.ABS.SG
VIIa/D
GEO (Me) pexi
momt’q’da.
(O)SV (me) PRON.1.DAT.SG
activated by -m- INFIX of
V under PLRPC
mo-m-t’q’-d-a V
PST-1.DAT.SG-
t’q-PASS-ABS
pexi N.ABS.SG IVc/C
m-DAT
Object
-a-ABS
Subject
GEOii Pexi movit’exe.
(O)SV (me) PRON.1.ERG.SG
activated by -v- INFIX of V
under PLRPC
mo-v-i.t’ex-e
PST-1.ERG.SG-
VERS-break-ABS
pexi N.ABS.SG IXb/B
Commentary:
In IT , as results from the respondents´ comments, the variant ITii is most frequently used. On the other hand, it activates
argument polysemy since it may also be interpreted intentionally if John did the breaking intentionally. Although the
intentional variant would be pragmatically improbable, it may be activated by the use of disjunct intenzionalmente that would
render the willed argument chain: Mi sono rotto la gamba volontariamente / intenzionalmente. Another intentional variant
may be Ho rotto la mia gamba. which would not sound quite common, but would be interpretable as a willed argument chain
most probably due to the pronoun mia.
Analysis of Sentence 10 The car broke its axle.
Lg Sentence WO Unintentional
Performer
Unwilled Action Specifier Coding
Marker/Alig
nment type
EN The car broke its
axle.
SVO the ART
car N.NOM.SG
its PRON POSS.3.SG in
OBL NP
broke-0 V.PST-1.SG axle
N.OBL.SG
VIId/D
DAN Bilens hjul gik i
stykker.
SV bilens N.GEN.SG gik V
go-PST.3.SG
i PREP
stykker OBL.PL
‘went into pieces’
hjul
N.NOM.SG
VIIa/D
98
SW Bilens hjul gick
sönder.
SV bilens N.GEN.SG gick V
go-PST.3.SG
sönder ADJ
hjul
N.NOM.SG
VIIa/D
IT Si è rotto l’asse
della macchina.
VS della PREP+ART
macchina N.GEN.SG
si REFL PART
é AUX
be-3.SG
rott-o
break-PST
PTCP.M.SG
l’asse
ART+N.NO
M.SG
VIII/E +
REFL PART
SP Al coche se le ha
roto la rueda.
SVO al ART
coche N.OBL.SG
le PRON.3.DAT.SG
se REFL PART.3.SG
h-a AUX
have-3.SG
rot-o
break-PST PTCP.
M.SG
se REFL PART.3.SG
for unplanned
occurrence
la ART
rueda
N.NOM.SG
V/C
(N-DAT
Le- DAT
Se- REFL
PART)
PG O carro partiu o
eixo.
SVO o ART
carro N.NOM.SG
-iu (INFL; NOM, 3SG)
part-iu V
break-PST.3.SG
o ART
eixo
N.ACC.SG
IIa/A
SKi Auto si zlomilo
nápravu.
SVO aut-o N.NOM.SG
si PRON
REFL.1.DAT.SG
zlomi-l-o V
break-PST-N.3.SG
si PRON
REFL.1.DAT.SG
náprav-u N.
ACC.SG
IIb/A+REFL
SKii Autu sa zlomila
náprava.
OVS aut-u N.SG.DAT zlomi-l-a V
break-PST-F.3.SG
sa REFL PART
náprav-a N.
NOM.SG
V/C
LITH Automobiliui
sulūžo ašis.
OVS automobiliui
N.DAT.SG
sulūžo V
break-PST.3.SG
ašis
N.NOM.SG
IVa/C
BG Счупи се оста на
колата.
VS na PREP
kolata N+ART
OBL/GEN.SG
schupi V.3.SG
se REFL PART
os-ta
N.OBL.SG-
ART
VIII/E
+REFL
PART
HU Az autó kereke
eltört
‘Car wheel-its
broke.’
SV az ART
autó N.NOM.SG
eltör-t V
break-PST.3.SG
kerek-e N.
OBL.SG-
POSS.3
VIIc/D+POSS
ESSIVENES
S
activated by -
e suffix of
Specifier
BASQ Kotxeak bere
gurpila apurtu
zuen
‘Car his wheel
broke.’
SOV kotxeak N.ERG.SG
bere PRON.POSS.3.SG
z- ERG INFL PREFIX
of AUX under PLRPC
apurtu-V
z-u-en-ERG/ABS
AUX
ERG.SG-u-ABS.3.SG
gurpila
N.ABS.SG
IXa/B
PLRPC
z- ERG
Subject
-en- ABS
Object
GEO Mankanas lilvi
most’q’da.
OSV mankana-s N
mankana-DAT.SG
PLRPC
-s-DAT INFL INFIX of
V
mo-s-t’q’-d-a V
PST-3.DAT.SG- t’q’-
PASS-ABS
/Preverb-Ind.OBJ-
brake-PASSIV-
3SUBJ/
lilvi
N.ABS.SG
IVb/C
PASS verb
form (passive
indicated by -
d- infix)
-s- DAT
Subject
-a-ABS
Subject
Commentary:
In BASQ auxiliary z-u-en the ERG concord is activated by the prefix -z (Hualde 2003:222).
Only EN and PG allow of the NOM Subject frame. SK has a NOM variant of Subject, but it is accompanied with DAT
reflexive pronoun. BASQ ERG/NOM frame is possible while in GEO it is DAT/ABS only. As the respondent remarked the
sentence would sound funny in Ergative as if a car were animated and it broke its axle willingly. This note might support an
99
idea that in GEO ERG plays a certain role as cognitive indicator, although in most of the sampled sentences it is used as
coding marker of the Subject when combined with ABS Argument in Aorist only.
Analysis of Sentence 11 Centipedes grow their legs (at different points in their development).
Lg Sentence WO Unintentional Performer Unwilled Action Specifier Coding
Marker/Alig
nment type
EN Centipedes grow
their legs.
SVO centipedes N.NOM.PL
their PRON POSS.3.PL in
OBL NP
grow-0 V.PRS-
3.PL
legs
N.OBL.PL
I/A + POSS
PRON – DET
in OBL NP
DAN Tusindben får
ben.
SVO tusindben N.NOM. SG&PL får-0 V
grow-3. SG&PL
ben
N+ART.OBL.
SG&PL
I/A
SW Tusenfotingar
utvecklar ben .
SVO tusenfotingar N.NOM.PL utvecklar-0 V
grow-3.PL
ben N+ART
OBL.SG&PL
I/A
IT Le gambe dei
millepiedi
crescono.
SV dei PREP+ART
millepiedi N.GEN.PL
crescon-o V
grow-3.PL
le ART
gambe
N.F.NOM. PL
VIII/E
SP A los ciempiés les
crecen las
piernas.
OVS a los PREP+ART
ciempiés N.DAT.PL
las ART.DAT.3.PL
crecen-0 V
grow-3.PL
las ART
piernas
N.NOM.PL
VI/C + extra
ART DAT
PG Às centopeias
crescem-lhes
pernas.
‘To the centipedes
grow-to.them
legs.’
OVS ás ART
centopeias N.DAT.PL
lhes PRON.DAT.3PL
crescem V.3.PL
pernas
ART+N.NO
M.PL
VI/C + extra
PRON DAT
SK Stonožkám rastú
nohy (v rôznych
vývojových
štádiách).
OVS stonožk-ám N
leg-DAT.PL
rast-ú V
grow-3.PL
noh-y N
leg-NOM.PL
IVa/C
LITH Lūpakojams
išauga kojos.
OVS lūpakojams N.DAT.PL išauga V.3.PL kojos
N.NOM.PL
IVa/C
BG На стоножките
краката им
израстват.
OSV na PREP
stonozhkite N+ART.OBL-
PL
im PRON POSS.3.PL
izrastvat V.3.PL krakata
N+ART.
NOM.PL
VI/C+
PRON POSS
in NOM NP
HU A százlábúak
önmaguk
növesztik lábaikat.
SVO a ART
százlábúak N.NOM.PL
önmaguk Intensifier
‘they themselves’
növesztik V.3.PL lábaikat
0ART+N.
ACC.PL
I/C+Intensifie
r
BASQ Ehunzangoek bere
hankak hasten
ditu.
SOV ehunzangoek N.ERG.SG
bere PRON POSS.3.SG
-u ERG INFL SUFFIX of
AUX under PLRPC
hasten-V
d-it-u ERG/ABS
AUX
3.ABS.PL-it-
ERG.SG
hankak
N.ABS.PL
IXa/B
PLRPC
u- ERG
Subject
d- ABS
Object
GEO Cxrapexebs
pexebi ezrdebat.
OSV cxrapex-eb-s N
cxrapex-PL-DAT
e-zrdeba-t V
PASS-zrdeba-3.PL
pexebi
N.ABS.PL
IVb/C
Commentary:
In Basque case endings are specified according to Hualde (2003:176), auxiliaries (ibid:222).
In Georgian the respondent did not indicate the pluripersonal concord in this sentence.
ENG, DAN, SW and HU code Unintentional Performer in NOM and BASQ as ERG. PG, SK, LITH and GEO activate it as
DAT via inflection and BG and SP via preposition adding also an extra DAT pronoun (in BG it behaves as a possessive). In
PG an extra DAT pronoun is added, too.
100
Analysis of Sentence 12 John grows a beard in winter. (willed action)
Lg Sentence WO Permitter
(intentional)
Willed Action Resultant (part of
Permitter)
Coding
Marker/Alig
nment type
EN John grows a
beard in winter.
SVO John N.NOM.SG grow-s V.PRS-
3.SG
a ART
beard N.OBL.SG
I/A
DAN John lader
skægget stå om
vinteren.
SVO John N.NOM.SG lader V
Causative.3.SG
stå V INF
skægget
N+ART.OBL.SG
I/A +
pseudomodal
causative
AUX +LEX
SW John låter
skägget växa på
vintern.
SVO John N.NOM.SG låter V
Causative.3.SG
växa V.INF
skägget
N+ART; OBL.SG
I/A +
pseudomodal
causative
AUX+LEX
IT John si fa
crescere la barba
in inverno.
SVO John N.NOM.SG
si PRON REFL.
3.DAT.SG
fa V
Causative.3.SG
crescere V INF
si PRON REFL.
3.DAT.SG
la DEF ART F.SG
barba N.OBL.F.SG
IIb/A+
pseudomodal
causative
AUX +
si PRON
REFL DAT
SP John se deja
barba en
invierno.
SVO John N.NOM.SG
se PRON REFL.
3.DAT.SG
deja V.3.SG
se PRON
REFL.3.DAT.SG
barba N.OBL.SG IIb/A+LEX
se PRON
REFL/DAT
PG O João deixa
crescer a barba
no inverno.
‘João lets grow,
permits to grow.’
SVO o ART
João N.NOM.SG
deixa V
Causative.3.SG
crescer V INF
a ART
barba N.ACC.SG
IIa/A
+pseudomoda
l causative
AUX
SKi John si pestuje
bradu v zime.
SVO John N.SG.NOM
si PRON
REFL.DAT.SG
si PRON REFL.
DAT.SG
pestuje V.3.SG
brad-u N
beard-ACC.SG
IIb/A+LEX
si PRON
REFL.DAT
SKii John si necháva
narásť bradu v
zime.
SVO John-0 N
John-NOM.SG
si PRON
REFL.DAT.SG
si PRON REFL.
DAT.SG
necháva V
Causative.3.SG
narásť V INF
brad-u N
beard-ACC.SG
IIb/A+pseudo
modal
causative
AUX+
si PRON
REFL.DAT
LITH Jonas žiemą
užsiaugina
barzdą.
SVO Jonas N.NOM.SG užsiaugina V
REFL.3.SG
barzdą N.ACC.SG IIb/A + REFL
V
BG Джон си пуска
брада през
зимата.
SVO Dzhon N.NOM.SG
si PRON
REFL.DAT.SG
si PRON
REFL.DAT.SG
puska V.3.SG
brada 0ART+N.
OBL.SG
IIb/A+LEX si
PRON REFL
DAT
HU János télen
szakállat növeszt.
SVO János N.NOM.SG növeszt V.3.SG szakállat N.
ACC.SG
IIa/A
BASQ Johnek bizarra
hasten du
neguan.
SOV Johnek N.ERG.SG
-u ERG INFL SUFFIX
of AUX under PLRPC
hasten-V
d-u -ERG/ABS
AUX
3.ABS.SG-
3.ERG.SG
bizzarra
N.ABS.SG
IXa/B
PLRPC
u- ERG
Subject
d- ABS
Object(?)
GEO Džoni zamtarši
c’vers izrdis.
SVO Džoni N.ABS.SG
PLRPC
-s- ABS INFL
SUFFIX of V
i-zrdi-s V
VERS-zrdi-
3.ABS.SG
c’ver-s N
beard-DAT.SG
III/A LEX
-s ABS
Subject
101
References Brigitte, Bauer. 2000. Archaic Syntax in Indo-European, Berlin. New York: Mouton de Gruyter.
Černý, Václav. 1971. Some remarks on syntax and morphology of verb in Avar. Archiv Orientalní 39.
46-56.
Dryer, Matthew S. 2013. Order of Subject, Object and Verb. In Dryer, Matthew S. & Haspelmath,
Martin (eds.), The World Atlas of Language Structures Online. Leipzig: Max Planck Institute
for Evolutionary Anthropology. < http://wals.info/chapter/81> Accessed 2016-08-12.
Etxepare, Ricardo. 2003.Valency and Argument Structure in the Basque Verb. In Hualde, José
Ignacio & de Urbina, Jon Ortiz (eds.), A Grammar of Basque. Berlin: Mouton de Gruyter.
Haspelmath, Martin. 2005. Argument Marking in Ditransitive Alignment Types. Linguistic Discovery.
3(1). Dartmouth College. <http://journals.dartmouth.edu/cgi-bin/WebObjects/Journals.woa/1/
xmlpage/ 1 /article/2. > 1-21. Accessed 2016-8-15.
Haspelmath, Martin. 2001. The European Linguistic Area: Standard Average European. In Haspelmath,
Martin & König, Ekkehard & Oesterreicher, Wulf & Raible, Wolfgang (eds.), Language
Typology and Language Universals – Handbücher zur Sprach- und
Kommunikationswissenschaft. Berlin: Mouton de de Gruyter. 1492–1510.
Hualde, José Ignacio & de Urbina, Jon Ortiz (eds.). 2003. A Grammar of Basque. Berlin: Mouton de
Gruyter.
Iggesen, Oliver A. 2013. Number of Cases. In: Dryer, Matthew S. & Haspelmath, Martin (eds.), The
World Atlas of Language Structures Online. Leipzig: Max Planck Institute for Evolutionary
Anthropology. <http://wals.info/chapter/49> Accessed on 2016-08-21.
Janigová, Slávka. 2014. Coding versus cognitive indication of valency reading of a NP/VP/NP sequence
– a cross-language study. Ostrava Journal of English Philology. 6 (1). 7-29.
Körtvélyessy, Lívia (2015). Evaluative Morphology from a Cross-Linguistic Perspective. Newcastle
upon Tyne: Cambridge Scholars Publishing.
Millward, Celia M. & Hayes Mary. 2012. Biography of the English Language. Boston: Wadsworth
Cengage Learning.
Skalička, Vladimír. 2004. Typ češtiny. In Čermák, František & Čermák, Jan &, Čermák, Petr &,
Poeta, Claudio (eds.), Vladimír Skalička Souborné dílo, 475-536. Prague: Karolinum.
Testelec, Yakov G. 1988. Word order in Kartvelian languages. Siewierska, Anna (ed.), Constituent
Order in the Languages of Europe. Berlin-New York: Mouton de Gruyter.
Van Valin, Robert D., Jr. (2001). An Introduction to Syntax. Cambridge: Cambridge University Press.
102
Slávka Janigová
Department of English and American Studies
P.J. Šafárik University
Moyzesova 9
040 01 Košice
Slovakia
slavka.janigova@gmail.com
In SKASE Journal of Theoretical Linguistics [online]. 2016, vol. 13, no.3 [cit. 2016-12-19].
Available on web page http://www.skase.sk/Volumes/JTL33/pdf_doc/05.pdf. ISSN 1336-
782X.
103